Metraipolis › archive › dc_bf116fd02208

XYBench: Can LLMs Respond Pragmatically to Queries with Misconceptions? — arXiv:2609.06842

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.06842
License (as archived)
CC0-1.0
Archived
2026-10-16T12:45:21.064Z
SHA-256 of the archived text
f112534a41a4ab3424c550ff697845143e2dabda78e38122e8aa6682bdb883bb
Stable URL of this page
https://metraipolis.com/documento/dc_bf116fd02208

How to cite

XYBench: Can LLMs Respond Pragmatically to Queries with Misconceptions? — arXiv:2609.06842 — arXiv (en), https://arxiv.org/abs/2609.06842 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bf116fd02208 (Metraipolis archive dc_bf116fd02208, sha256 f112534a41a4ab3424c550ff697845143e2dabda78e38122e8aa6682bdb883bb)
BibTeX
@misc{dcbf116fd02208,
  title = {XYBench: Can LLMs Respond Pragmatically to Queries with Misconceptions? — arXiv:2609.06842},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.06842},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bf116fd02208; sha256 f112534a41a4ab3424c550ff697845143e2dabda78e38122e8aa6682bdb883bb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06842. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

When non-expert users ask LLMs for assistance, their queries can often have misconceptions (e.g., "How do I parse XML with regex?"). In such cases, often referred to as the XY-problem, LLMs must identify the misconception ("regex are fragile") and meaningfully direct the user toward a pragmatic solution that will address the root problem implicit in the request ("use an XML parser"). We introduce XYBench, a benchmark of 8,115 such queries, drawn from technical (StackOverflow/StackExchange) and everyday (WikiHow and a manually-curated subset) domains. We design an evaluation paradigm that assesses model responses along three criteria grounded in cooperative response theory: (a) presence and (b) emphasis on pragmatic solutions, and (c) identification of misconceptions. Our experiments show that even the strongest LLMs predominantly answer the literal request (0.75--0.92) and far less often the intended one (0.33--0.71), while substantially lagging behind humans at identifying misconceptions (at most 63% vs. 79--90%). Further, models overwhelmingly prefer pragmatic responses in a multiple choice setting yet consistently fail to generate them. Oracle ablation experiments show that providing explicit user intent at generation time helps; however a large gap remains, suggesting pragmatic redirection is a fundamentally underdeveloped capability in current LLMs.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.06842 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T12:30:25.068Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).